A congestion prediction-based navigation method, device and electronic equipment

Abstract

The application belongs to the field of navigation, and discloses a navigation method and device based on congestion prediction and electronic equipment, the method comprising: constructing an agent-based model according to traffic flow data; identifying a key area by using a traffic network model; simulating traffic flow by using the agent-based model, and predicting a congestion area according to the traffic flow simulation result and the key area; and generating a navigation route according to navigation information and the congestion area. The agent-based model constructed by combining traffic flow data and the traffic network model can comprehensively analyze and simulate traffic flow dynamics, thereby accurately identifying a key area and predicting a potential congestion area. Not only the behavior of vehicles as individuals and the structural features of the overall road network are considered separately, but also the analysis results of the two are fused to more comprehensively and accurately predict the congestion area, thereby improving the accuracy of congestion prediction, and by generating an optimized navigation route, congestion is effectively avoided, and the travel efficiency of users is improved.

Description

Technical Field

[0001] This invention relates to the field of navigation, and more specifically, to a navigation method, apparatus, electronic device, and computer-readable storage medium based on congestion prediction. Background Technology

[0002] With the acceleration of urbanization, urban traffic congestion has become increasingly serious, significantly impacting people's travel efficiency and quality of life. Traditional navigation systems mainly rely on real-time traffic data to provide optimal routes, but these systems often ignore the dynamic changes and complexity of traffic flow. To more effectively solve the traffic congestion problem, a navigation method capable of predicting future traffic conditions is needed.

[0003] Existing traffic prediction methods are mainly divided into two categories: models based on individual vehicle behavior and models based on the overall road network structure. Models based on individual vehicle behavior, such as vehicle following models and lane-changing models, primarily focus on the interactions between vehicles and individual driving behaviors. These models can simulate the micro-behavior of vehicles in traffic flow relatively well, but often struggle to handle the complexity of large-scale traffic networks. On the other hand, models based on the overall road network structure, such as network flow models and traffic assignment models, focus on analyzing the topology of the road network and the overall distribution of traffic flow. These models provide the overall trend of traffic flow at a macro level, but they do not adequately consider the behavior of individual vehicles.

[0004] However, relying solely on the analysis of individual vehicle behavior or the overall road network structure makes it difficult to predict traffic congestion comprehensively and accurately, resulting in existing navigation technologies being unable to accurately avoid congested areas. Summary of the Invention

[0005] To address the problem that existing navigation technologies are unable to accurately avoid congested areas due to the randomness of individual vehicle behavior and the complexity of road network structures, this invention provides a navigation method based on congestion prediction.

[0006] A navigation method based on congestion prediction includes:

[0007] Traffic flow data for the corresponding area is obtained based on the input navigation information, and a proxy-based model is constructed based on the traffic flow data.

[0008] A traffic network model is obtained based on the navigation information, and key areas are identified using the traffic network model.

[0009] Traffic flow is simulated using the agent-based model, and congestion areas are predicted based on the traffic flow simulation results and the key areas.

[0010] A navigation route is generated based on the navigation information and the congested area.

[0011] Optionally, constructing an agent-based model based on the traffic flow data includes:

[0012] Identify the relationships between traffic participants based on the traffic flow data;

[0013] The agent-based model is constructed based on the relationships between each of the traffic participants;

[0014] The type of each traffic participant is identified based on the traffic flow data, and behavioral rules for each agent in the agent-based model are set according to the type.

[0015] Optionally, simulating traffic flow using the agent-based model includes:

[0016] In the agent-based model, a first agent is generated;

[0017] The navigation information is used to identify the type of traffic participant to which the current user belongs, and the behavior rules of the first agent are set according to the type of traffic participant to which the current user belongs;

[0018] Set at least one initial navigation route for the first agent based on the navigation information;

[0019] The traffic flow simulation results are obtained by simulating the traffic flow of the first agent traveling on each of the initial navigation routes using the agent-based model.

[0020] Optionally, the step of predicting congested areas based on the traffic flow simulation results and the key areas includes:

[0021] Based on the simulation results of each traffic flow, areas where the number of agents clustering exceeds a first threshold are marked as dense areas;

[0022] The densely populated areas that overlap with the key areas are identified as the congested areas.

[0023] Optionally, before simulating the traffic flow of the first agent traveling on each of the initial navigation routes using the agent-based model, the method further includes:

[0024] A second agent is added to the agent-based model, and the number of the second agents is less than a second threshold.

[0025] Optionally, obtaining the traffic network model includes:

[0026] The navigation information is obtained to determine the navigation area, and the traffic routes in the navigation area and the historical traffic flow data corresponding to the traffic routes are obtained.

[0027] The traffic network model is generated based on the traffic routes, and the attributes of each edge in the traffic network model are determined based on the historical traffic flow data.

[0028] Optionally, identifying key areas using the traffic network model includes:

[0029] Key edges are determined based on the attributes of each edge in the traffic network model;

[0030] Calculate the betweenness centrality of each node in the traffic network model, and determine the key nodes based on the betweenness centrality;

[0031] The key region is determined based on the key edge and the key node.

[0032] A navigation device based on congestion prediction includes:

[0033] The first acquisition module is used to acquire traffic flow data of the corresponding area based on the input navigation information, and to construct an agent-based model based on the traffic flow data;

[0034] The second acquisition module is used to acquire a traffic network model based on the navigation information and to identify key areas using the traffic network model.

[0035] The simulation module is used to simulate traffic flow using the agent-based model and predict congestion areas based on the traffic flow simulation results and the key areas.

[0036] The navigation module is used to generate a navigation route based on the navigation information and the congested area.

[0037] An electronic device, comprising:

[0038] A processor and a memory, the memory being used to store at least one instruction, which, when loaded and executed by the processor, implements the congestion prediction-based navigation method as described in any of the preceding embodiments.

[0039] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the congestion prediction-based navigation method as described in any of the preceding claims.

[0040] The congestion prediction-based navigation method provided in this invention obtains traffic flow data for a corresponding area based on input navigation information and constructs an agent-based model based on the traffic flow data. It then acquires a traffic network model based on the navigation information and uses this model to identify key areas. The agent-based model is used to simulate traffic flow, and congested areas are predicted based on the simulation results and key areas. Finally, a navigation route is generated based on the navigation information and congested areas. This invention, by combining the agent-based model constructed from traffic flow data with the traffic network model, can comprehensively analyze and simulate traffic flow dynamics, thereby accurately identifying key areas and predicting potential congestion areas. It not only considers the individual behavior of vehicles and the structural characteristics of the overall road network but also integrates the analysis results of both to more comprehensively and accurately predict congested areas, improving the accuracy of congestion prediction. Furthermore, by generating optimized navigation routes, it effectively avoids congestion, improving user travel efficiency and experience. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 A flowchart illustrating a congestion prediction-based navigation method provided in an embodiment of the present invention;

[0043] Figure 2 A flowchart illustrating another congestion prediction-based navigation method provided in an embodiment of the present invention;

[0044] Figure 3 This is a schematic diagram of a congestion prediction-based navigation device provided in an embodiment of the present invention. Detailed Implementation

[0045] To better understand the technical solution of the present invention, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0046] It should be understood that the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0047] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0048] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0049] Existing traffic prediction methods are mainly divided into two categories: models based on individual vehicle behavior and models based on the overall road network structure. Models based on individual vehicle behavior, such as vehicle following models and lane-changing models, primarily focus on the interactions between vehicles and individual driving behaviors. These models can simulate the micro-behavior of vehicles in traffic flow relatively well, but often struggle to handle the complexity of large-scale traffic networks. On the other hand, models based on the overall road network structure, such as network flow models and traffic assignment models, focus on analyzing the topology of the road network and the overall distribution of traffic flow. These models provide the overall trend of traffic flow at a macro level, but they do not adequately consider the behavior of individual vehicles.

[0050] However, relying solely on the analysis of individual vehicle behavior or the overall road network structure makes it difficult to comprehensively and accurately predict traffic congestion, resulting in existing navigation technologies being unable to accurately avoid congested areas. Therefore, this invention provides a navigation method based on congestion prediction to solve the above problems.

[0051] Please refer to Figure 1 The flowchart of a congestion prediction-based navigation method provided in an embodiment of the present invention includes the following steps:

[0052] Step S01: Obtain traffic flow data for the corresponding area based on the input navigation information, and construct an agent-based model based on the traffic flow data.

[0053] In this embodiment, the agent-based model is a simulation technique that simulates the dynamic changes in traffic flow by creating agents representing actual traffic participants (such as vehicles and pedestrians) and assigning them specific behavioral rules. This model can meticulously reflect the impact of individual behavior on overall traffic flow, making the traffic flow simulation closer to reality. When a user inputs navigation information, the system can use this information to determine the user's navigation route and location, thereby accurately locating a specific geographical area. This allows for targeted collection and analysis of real-time traffic flow data for that area. The collected data includes vehicle speed, traffic volume, accident information, etc., which provides the foundation for building the agent-based model. This embodiment not only provides a deep understanding of the current traffic situation but also predicts future traffic trends, providing the navigation system with more accurate congestion predictions and route planning, thus optimizing the user's travel experience.

[0054] In some embodiments, the method of acquiring traffic flow data may include:

[0055] 1) By deploying sensor networks, such as cameras and radar, these devices can capture real-time information on vehicle position, speed, and trajectory at and around intersections.

[0056] 2) For cities that have open traffic data, they can directly obtain real-time traffic data for specific intersections from the traffic management center, which may include key information such as vehicle flow and traffic signal status.

[0057] 3) Utilize third-party map service platforms, such as Google Maps or Baidu Maps, which provide rich real-time traffic data, including congestion and estimated travel time.

[0058] The aforementioned methods of traffic flow data collection can provide comprehensive and detailed data support for building accurate traffic flow models, which helps improve the efficiency and accuracy of traffic management and navigation services.

[0059] For example, as mentioned in step S01, obtaining traffic flow data for the corresponding area based on the input navigation information and constructing an agent-based model based on the traffic flow data can be specifically described as follows:

[0060] When a user inputs their destination and other relevant navigation parameters through a navigation device or application, the system calculates one or more possible initial routes based on this information, identifying the specific areas these routes traverse, such as city streets and highways. The system then collects real-time traffic flow data for these areas from data sources including traffic management centers, vehicle sensors, cameras, and mobile applications, including vehicle speed, traffic volume, accident and construction information. Based on this traffic flow data, the system can construct an agent-based model, where each agent represents a traffic participant, such as a vehicle or pedestrian, and sets behavioral rules for each agent based on the collected traffic flow data, such as driving speed, lane-changing frequency, and parking habits. By simulating the interaction and movement of these agents in the traffic network, traffic flow trends can be predicted, thus providing the navigation system with accurate congestion predictions and optimized route suggestions.

[0061] Step S02: Obtain a traffic network model based on navigation information, and use the traffic network model to identify key areas.

[0062] In this embodiment, the traffic network model constructed by analyzing navigation information can describe the topology of the road network and the dynamic characteristics of traffic flow in detail. The traffic network model can identify key areas in the traffic network, such as road sections that are frequently congested, accident-prone points, or intersections where traffic signals need to be optimized. This can provide early warnings of potential traffic problems and offer drivers more accurate navigation routes, thereby improving the overall traffic conditions and travel experience.

[0063] In some embodiments, to accurately predict future traffic conditions and identify potential congestion points and traffic bottlenecks, a traffic network model can be established by analyzing historical traffic flow data, i.e., as mentioned in step S02, obtaining the traffic network model, which may specifically include:

[0064] Step S021: Obtain navigation information to determine the navigation area, and obtain the traffic routes in the navigation area and the historical traffic flow data corresponding to the traffic routes.

[0065] In this embodiment, key information is extracted from the navigation system to define specific geographical areas, and detailed information and historical data of traffic routes within the navigation area are further collected to provide basic data for building and analyzing traffic network models. This includes, but is not limited to, road layout, traffic signals, traffic signs, and historical data such as traffic flow, speed, and accident records on these routes. This allows the traffic network model built based on the historical traffic flow data to reflect the actual situation and historical trends of traffic flow, thereby identifying key areas in the traffic network, such as road sections where congestion or accidents frequently occur.

[0066] Step S022: Generate a traffic network model based on traffic routes, and determine the attributes of each edge in the traffic network model based on historical traffic flow data.

[0067] In this embodiment, an abstract yet fully functional traffic network model is constructed by transforming actual traffic routes into edges and nodes in a mathematical or computational model. Each edge represents a road segment, and its attributes, such as capacity, speed limit, and historical average flow rate, are set based on long-term collected historical traffic flow data to accurately reflect the traffic characteristics and behavior patterns of each road segment in the traffic network model. This allows for the identification of road segments prone to congestion during specific time periods, or areas frequently experiencing traffic disruptions due to accidents.

[0068] Based on the above embodiments, in some embodiments, to further improve the accuracy of key area identification, it can be achieved by determining the key edges and key nodes in the traffic network model, that is, as mentioned in step S02, using the traffic network model to identify key areas, which specifically may include:

[0069] Step S023: Determine the key edges based on the attributes of each edge in the traffic network model.

[0070] In this embodiment, by evaluating specific attributes of each edge (representing a road segment) in the traffic network model, such as traffic volume, speed, and historical congestion data, road segments that play a critical role in traffic flow are identified. These attributes indicate the importance of each edge in the overall traffic network and its potential congestion risk; when traffic pressure is high, critical edges have a higher congestion risk than ordinary edges.

[0071] Step S024: Calculate the betweenness centrality of each node in the traffic network model, and determine the key nodes based on the betweenness centrality.

[0072] In this embodiment, the control and influence of each node in traffic flow are assessed by measuring the frequency of each node's occurrence on all shortest paths in the transportation network. Nodes with high betweenness centrality mean that their geographical location within the network is crucial for connecting different areas; they are often convergence or dispersal points of traffic flow. If a node has high betweenness centrality, it means that many node pairs need to pass through this node for communication or connection, thus playing a "bridge" or "mediator" role in the network. The purpose of identifying these key nodes is to mark areas that have a significant impact on traffic flow in traffic management and navigation planning, as these nodes are more likely to become bottlenecks or critical control points in traffic congestion, accidents, or other disruptive events. In this way, key areas of urban traffic can be identified more effectively.

[0073] In some embodiments, the calculation of betweenness centrality can be expressed as:

[0074]

[0075] Where B(v) is the betweenness centrality of node (v), σ st σ is the number of all shortest paths from node(s) to node(t). st (v) represents the number of shortest paths from (s) to (t) that pass through node (v).

[0076] Step S025: Determine the key region based on the key edges and key nodes.

[0077] In this embodiment, critical areas crucial to the operation of the entire transportation network are defined by identifying key edges (i.e., road segments) carrying high traffic volume or of strategic significance, and key nodes (such as intersections or transportation hubs) that play a central role in traffic flow. This allows resources and attention to be focused on the parts that have the greatest impact on traffic flow and efficiency, as these areas are often the focus of traffic congestion, accident occurrences, or traffic management measures. By identifying these critical areas, potential congestion areas can be predicted more accurately, providing navigation systems with accurate congestion predictions and optimized route suggestions.

[0078] Step S03: Simulate traffic flow using an agent-based model, and predict congested areas based on the traffic flow simulation results and key areas.

[0079] In this embodiment, real-world traffic flow is simulated by creating agents representing actual traffic participants (such as vehicles and pedestrians) and assigning them behavioral rules and interaction mechanisms in a simulated environment. Simultaneously, the traffic flow simulation results are combined with key areas identified using a traffic network model to predict congestion areas. This considers not only the individual behavioral characteristics of vehicles, such as speed, acceleration, and lane-changing behavior, but also the structural characteristics of the overall road network, such as road capacity, traffic signals, and network topology. By fusing the analysis results of individual behavior and network structure, the occurrence and development of traffic congestion can be predicted more comprehensively and accurately. This allows for optimal route selection during navigation, helping users avoid potential congestion areas, thereby reducing travel time, improving travel efficiency, and optimizing the individual travel experience.

[0080] In some embodiments, to ensure the efficiency of congestion area prediction and to increase the coverage area of ​​the predicted congestion areas, key areas and locations where a large number of traffic participants gather in the traffic flow simulation results can be identified as congestion areas.

[0081] Step S04: Generate a navigation route based on navigation information and congested areas.

[0082] In this embodiment, by utilizing navigation information and congested areas to analyze road networks and traffic conditions, it is possible to avoid road sections that are expected to be congested and choose relatively smooth routes, thereby reducing users' travel time and improving travel efficiency. This provides users with more intelligent and personalized navigation services, ensuring that users can reach their destination quickly and safely, while reducing the inconvenience and time wastage caused by traffic congestion.

[0083] Based on the above technical solution, this invention obtains traffic flow data for the corresponding area based on input navigation information and constructs an agent-based model based on the traffic flow data; it obtains a traffic network model based on navigation information and uses the traffic network model to identify key areas; it simulates traffic flow using the agent-based model and predicts congested areas based on the traffic flow simulation results and key areas; and it generates navigation routes based on navigation information and congested areas. This invention, by combining the agent-based model constructed from traffic flow data with the traffic network model, can comprehensively analyze and simulate traffic flow dynamics, thereby accurately identifying key areas and predicting potential congested areas. It not only considers the behavior of vehicles as individuals and the structural characteristics of the overall road network separately, but also integrates the analysis results of both to more comprehensively and accurately predict congested areas, improving the accuracy of congestion prediction. Furthermore, by generating optimized navigation routes, it effectively avoids congestion, improving user travel efficiency and experience.

[0084] Please refer to Figure 2 This is a flowchart illustrating another congestion prediction-based navigation method provided in an embodiment of the present invention. In some embodiments, to improve the adaptability and accuracy of the agent-based model and thus more realistically reflect traffic flow dynamics, the following can also be performed: Figure 2 The steps shown are as follows:

[0085] Step S11: Identify the relationships between traffic participants based on traffic flow data.

[0086] In this embodiment, by analyzing traffic flow data, such as vehicle position, speed, and acceleration, the relative positions and interactions of different traffic participants (such as vehicles, pedestrians, and bicycles) within the traffic flow are determined. This enables the system to understand and simulate the complex dynamics of traffic flow, such as following, yielding, and overtaking behaviors between vehicles. By identifying these interrelationships, the changing trends of traffic flow can be predicted more accurately, thus providing crucial information for building agent-based models. Such models can simulate the individual behaviors of traffic participants and their impact on overall traffic flow, thereby improving the accuracy of traffic congestion prediction and the effectiveness of navigation route planning, optimizing traffic management, and enhancing road utilization efficiency.

[0087] In some embodiments, the relationships between traffic participants may include following relationships, cooperative relationships, and competitive relationships; based on this, the step S11, which involves identifying the relationships between traffic participants based on traffic flow data, can specifically be as follows:

[0088] Video analytics technology is used to track the historical trajectory of each traffic participant and determine whether a following relationship exists between multiple traffic participants based on the historical trajectory.

[0089] Time series analysis is used to analyze the changing trends of traffic flow data, and the changing trends are used to determine whether there are cooperative or competitive relationships among multiple traffic participants.

[0090] In this embodiment, the relationships between traffic participants are identified based on traffic flow data. Sociophysical principles are applied to analyze collected vehicle behavior data, revealing the interaction mechanisms between individuals within a group. Specifically, firstly, video analytics or sensor data is used to automatically track vehicle trajectories to determine if a vehicle is following another. In-depth analysis of the spatiotemporal relationship between following vehicles and the vehicle in front quantifies the patterns and frequency of following behavior. Secondly, time-series data is used to observe trends in traffic flow, which help infer cooperative or competitive relationships within vehicle group behavior. For example, before congestion occurs, vehicles may exhibit behaviors such as accelerating to compete for entry into faster lanes, or in certain situations, vehicles may exhibit cooperative behavior, such as alternating lane changes. Through these analyses, a complex network of interactions among traffic participants can be constructed, providing a scientific basis for understanding and predicting traffic flow dynamics.

[0091] This embodiment provides a deep understanding of individual behavior and group dynamics in traffic flow, thus offering a scientific basis for traffic management and planning. By analyzing vehicle following behavior and group cooperation or competition, the occurrence and development of traffic congestion can be predicted more accurately, optimizing traffic signal control and improving road utilization efficiency. Furthermore, this embodiment helps identify potential problems in traffic flow, such as congestion exacerbated by excessive competition, allowing for targeted mitigation measures, thereby improving the safety and smoothness of the traffic system and providing drivers with a faster and more comfortable travel experience.

[0092] Step S12: Construct an agent-based model based on the relationships between each traffic participant.

[0093] In this embodiment, the agent-based model is a powerful computational model that allows researchers to explore how individual-level decisions and behaviors form group phenomena at a macro level by simulating individual-level decisions and behaviors. In the agent-based model, each agent, such as a vehicle, bicycle, or pedestrian, possesses a certain degree of autonomy, enabling it to make independent decisions based on preset rules and interact with other agents.

[0094] Step S13: Identify the type of each traffic participant based on traffic flow data, and set the behavior rules for each agent in the agent-based model according to the type.

[0095] In this embodiment, traffic flow data is analyzed to distinguish different types of traffic participants, such as private cars, buses, trucks, and motorcycles. Each type of participant has unique behavioral characteristics, such as driving speed, acceleration, and route selection. In the agent-based model, these characteristics are transformed into specific behavioral rules to guide each agent in simulating its corresponding behavior. This makes the model more realistically reflect the actual performance of different types of vehicles in traffic flow, improving the accuracy of traffic simulation and the effectiveness of prediction. By setting appropriate behavioral rules for different types of traffic participants, the occurrence, development, and dissipation of traffic congestion can be predicted more accurately, thereby providing more targeted strategies for traffic management and planning, optimizing traffic flow, reducing congestion, and improving road use efficiency.

[0096] Step S14: Obtain a traffic network model based on navigation information, and use the traffic network model to identify key areas.

[0097] Step S15: Generate the first agent in the agent-based model.

[0098] Step S16: Identify the type of traffic participant to which the current user belongs based on the navigation information, and set the behavior rules of the first agent based on the type of traffic participant to which the current user belongs.

[0099] In this embodiment, the user's navigation information is first analyzed to determine the type of traffic participant the user belongs to, such as a car, electric vehicle, or pedestrian, in order to predict the user's possible driving behavior and route selection. Then, based on the type of traffic participant, corresponding behavioral rules are set for the first agent representing the user, such as different speed limits, driving habits, or reactions to traffic signals. This personalized setting ensures that the model can more accurately simulate the user's actual driving behavior, thereby making the traffic flow simulation results more realistic and reliable.

[0100] In this embodiment, a first agent representing the navigation-initiating traffic participant (such as a vehicle or pedestrian) is created in the computational model of simulated traffic flow. This first agent is assigned a series of behavioral rules and characteristics, such as speed preference, acceleration capability, and response to traffic signals, enabling it to simulate the behavior of real traffic participants in the simulated environment. This allows the dynamic behavior of the navigation-initiating traffic participant to be introduced into the traffic flow simulation, thereby more accurately predicting and analyzing the situations that the navigation-initiating traffic participant may encounter in traffic flow, including traffic congestion, vehicle distribution, and driving safety.

[0101] Step S17: Set at least one initial navigation route for the first agent based on the navigation information.

[0102] In this embodiment, by utilizing the user's starting point, destination, and other relevant navigation data in the agent-based model, one or more possible driving paths are determined for the first agent representing the user. This reproduces the user's actual driving situation in the simulation environment, ensuring that the simulation results reflect real-world traffic flow dynamics. By setting an initial navigation route for the first agent, the model can predict potential traffic conditions encountered on specific routes, such as congestion, smooth traffic, or accidents. This provides the user with more accurate route suggestions and travel time estimates, optimizing the user's travel experience, reducing travel time, and improving the practicality and accuracy of the navigation system.

[0103] Step S18: Simulate the traffic flow of the first agent traveling on each initial navigation route using an agent-based model to obtain the traffic flow simulation results.

[0104] In this embodiment, by placing the first agent on different routes and simulating its driving process according to the set behavior rules and traffic conditions, and then observing and recording the driving performance of the first agent under various conditions, including speed, frequency of encountering congestion, and travel time, detailed traffic flow data can be generated to predict traffic conditions on different routes, thereby providing more accurate navigation suggestions for users' actual travel.

[0105] In some embodiments, to predict potential congestion during peak hours, a second agent can be added to the agent-based model to simulate the increase in the number of vehicles on the road during peak hours. That is, before executing step S18, the following steps can also be performed:

[0106] In the agent-based model, a second agent is added, and the number of second agents is less than a second threshold.

[0107] In this embodiment, the second threshold is the maximum number of second agents that can be added. By setting the second threshold, the system can prevent the agent-based model from stopping due to an excessive number of agents.

[0108] Step S19: Based on the simulation results of each traffic flow, areas where the number of agents clustering exceeds the first threshold are marked as dense areas.

[0109] In this embodiment, potential traffic congestion points are identified by evaluating the distribution density of agents (representing vehicles or other traffic participants) in simulated traffic flow. A first threshold is the minimum number of agents clustered in a dense area. When the number of agents in an area exceeds the first threshold, it indicates that the traffic flow in that area is high and there may be a risk of congestion. The purpose of marking these areas as dense areas is to provide real-time traffic information in the navigation system, helping drivers proactively avoid these congested areas when planning routes, thereby optimizing travel routes and reducing travel delays.

[0110] Step S110: Identify the dense areas that overlap with the key areas as congested areas.

[0111] In this embodiment, actual congested areas are determined by comparing densely populated areas identified in traffic flow simulation with predefined key areas. These areas are then precisely marked in the navigation system as places with high traffic volume, slow-moving traffic, or potential traffic problems. In this way, the navigation system can automatically avoid these identified congested areas when planning routes, providing users with faster and smoother alternative routes, thus optimizing the driver's travel experience and reducing time losses caused by traffic congestion.

[0112] Step S111: Generate a navigation route based on navigation information and congested areas.

[0113] Based on the above technical solutions, this invention, by constructing an agent-based model, can accurately simulate the behavior of individual traffic participants and their interrelationships, thereby more realistically reflecting traffic flow dynamics. This method improves the model's adaptability and accuracy by identifying traffic participant types and setting behavioral rules for them. Furthermore, by simulating traffic flow on different initial navigation routes, potential congestion areas can be predicted and identified, providing drivers with route options to avoid these areas. This effectively optimizes route planning, reduces congestion, improves travel efficiency, and provides users with more accurate and personalized navigation services.

[0114] Please refer to Figure 3 This is a schematic diagram of a congestion prediction-based navigation device provided in an embodiment of the present invention. The congestion prediction-based navigation device includes:

[0115] The first acquisition module 100 is used to acquire traffic flow data of the corresponding area based on the input navigation information, and to construct an agent-based model based on the traffic flow data.

[0116] The second acquisition module 200 is used to acquire a traffic network model based on navigation information and to identify key areas using the traffic network model.

[0117] The simulation module 300 is used to simulate traffic flow using an agent-based model and predict congestion areas based on the traffic flow simulation results and key areas.

[0118] The navigation module 400 is used to generate navigation routes based on navigation information and congested areas.

[0119] Based on the above embodiments, in one specific embodiment, the first acquisition module 100 may be used for:

[0120] Identify the relationships between traffic participants based on traffic flow data;

[0121] Construct an agent-based model based on the relationships between each traffic participant;

[0122] Identify the type of each traffic participant based on traffic flow data, and set behavioral rules for each agent in the agent-based model according to the type.

[0123] Based on the above embodiments, in one specific embodiment, the simulation module 300 can be specifically used for:

[0124] Generate the first agent in the agent-based model;

[0125] Identify the type of traffic participant to which the current user belongs based on navigation information, and set the behavior rules for the first agent based on the type of traffic participant to which the current user belongs;

[0126] Set at least one initial navigation route for the first agent based on the navigation information;

[0127] Traffic flow simulation results were obtained by simulating the traffic flow of the first agent traveling on each initial navigation route using an agent-based model.

[0128] Based on the above embodiments, in one specific embodiment, the simulation module 300 can be specifically used for:

[0129] Based on the simulation results of each traffic flow, areas where the number of agents clustering exceeds a first threshold are marked as dense areas;

[0130] Densely populated areas that overlap with key areas are identified as congested areas.

[0131] Based on the above embodiments, in one specific embodiment, the simulation module 300 can also be used for:

[0132] In the agent-based model, a second agent is added, and the number of second agents is less than a second threshold.

[0133] Based on the above embodiments, in one specific embodiment, the second acquisition module 200 may be used for:

[0134] Obtain navigation information to determine the navigation area, and obtain traffic routes within the navigation area as well as historical traffic flow data corresponding to those routes.

[0135] A traffic network model is generated based on traffic routes, and the attributes of each edge in the traffic network model are determined based on historical traffic flow data.

[0136] Based on the above embodiments, in one specific embodiment, the second acquisition module 200 may be used for:

[0137] Key edges are determined based on the attributes of each edge in the traffic network model;

[0138] Calculate the betweenness centrality of each node in the traffic network model, and determine the key nodes based on the betweenness centrality;

[0139] Determine the critical region based on the critical edge and critical node.

[0140] This embodiment provides an electronic device, including a processor and a memory. The memory is used to store at least one instruction. When the instruction is loaded and executed by the processor, it implements the above-described congestion prediction-based navigation method. Its execution method and beneficial effects are similar and will not be described again here.

[0141] This invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the above-described navigation method based on congestion prediction. Its execution method and beneficial effects are similar and will not be described again here.

[0142] It should be noted that although the steps are described in a specific order above, it does not mean that the steps must be executed in the above specific order. In fact, some of these steps can be executed concurrently, or even in a different order, as long as the required function can be achieved.

[0143] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A navigation method based on congestion prediction, characterized in that, include: Traffic flow data for the corresponding area is obtained based on the input navigation information, and a proxy-based model is constructed based on the traffic flow data. A traffic network model is obtained based on the navigation information, and key areas are identified using the traffic network model. Traffic flow is simulated using the agent-based model, and congestion areas are predicted based on the traffic flow simulation results and the key areas. A navigation route is generated based on the navigation information and the congested area.

2. The navigation method according to claim 1, characterized in that, The step of constructing an agent-based model based on the traffic flow data includes: Identify the relationships between traffic participants based on the traffic flow data; The agent-based model is constructed based on the relationships between each of the traffic participants; The type of each traffic participant is identified based on the traffic flow data, and behavioral rules for each agent in the agent-based model are set according to the type.

3. The navigation method according to claim 2, characterized in that, The simulation of traffic flow using the agent-based model includes: In the agent-based model, a first agent is generated; The navigation information is used to identify the type of traffic participant to which the current user belongs, and the behavior rules of the first agent are set according to the type of traffic participant to which the current user belongs; Set at least one initial navigation route for the first agent based on the navigation information; The traffic flow simulation results are obtained by simulating the traffic flow of the first agent traveling on each of the initial navigation routes using the agent-based model.

4. The navigation method according to claim 3, characterized in that, The prediction of congested areas based on the traffic flow simulation results and the key areas includes: Based on the simulation results of each traffic flow, areas where the number of agents clustering exceeds a first threshold are marked as dense areas; The densely populated areas that overlap with the key areas are identified as the congested areas.

5. The navigation method according to claim 3, characterized in that, Before simulating the traffic flow of the first agent traveling on each of the initial navigation routes using the agent-based model, the method further includes: A second agent is added to the agent-based model, and the number of the second agents is less than a second threshold.

6. The navigation method according to claim 1, characterized in that, The acquisition of the traffic network model includes: The navigation information is obtained to determine the navigation area, and the traffic routes in the navigation area and the historical traffic flow data corresponding to the traffic routes are obtained. The traffic network model is generated based on the traffic routes, and the attributes of each edge in the traffic network model are determined based on the historical traffic flow data.

7. The navigation method according to claim 6, characterized in that, The identification of key areas using the traffic network model includes: Key edges are determined based on the attributes of each edge in the traffic network model; Calculate the betweenness centrality of each node in the traffic network model, and determine the key nodes based on the betweenness centrality; The key region is determined based on the key edge and the key node.

8. A navigation device based on congestion prediction, characterized in that, include: The first acquisition module is used to acquire traffic flow data of the corresponding area based on the input navigation information, and to construct an agent-based model based on the traffic flow data; The second acquisition module is used to acquire a traffic network model based on the navigation information and to identify key areas using the traffic network model. The simulation module is used to simulate traffic flow using the agent-based model and predict congestion areas based on the traffic flow simulation results and the key areas. The navigation module is used to generate a navigation route based on the navigation information and the congested area.

9. An electronic device, characterized in that, include: A processor and a memory, the memory being used to store at least one instruction, which, when loaded and executed by the processor, implements the congestion prediction-based navigation method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the congestion prediction-based navigation method as described in any one of claims 1-7.

Images